AI-Generated Distributed Algorithms: Paving the Way for More Dependable Distributed Systems

Fault-tolerant distributed algorithms contribute to the reliability and resilience of modern computation due to their fault-tolerant properties. Moreover, the increasing need for computing power demands building efficient systems. However, the design of distributed algorithms is often laborious, complex, and time-consuming. We address this challenge by proposing an artificial intelligence agent capable of automatically generating distributed algorithms. In this paper, we focus on the study of binary consensus algorithms. Our preliminary results show the generation of a more efficient algorithm.